The change in biotic and abiotic soil components influenced by paddy soil microbial fuel cells ...

26 Pages·2017·1.97 MB·English

Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 8 February 2018 doi:10.20944/preprints201802.0067.v1
Peer-reviewed version available at Journal of Soils and Sediments 2018; doi:10.1007/s11368-018-2024-1
1 Article
The change in biotic and abiotic soil components influenced
2
by paddy soil microbial fuel cells loaded with various
3
resistances
4
5
6 Williamson Gustave ab, Zhao-Feng Yuan ab, Raju Sekar c, Yu-Xiang Ren a, Hu-Cheng Changa
7 and Zheng Chen a*
8
9
10 a Department of Environmental Science, Xi'an Jiaotong-Liverpool University, Suzhou, Jiangsu
11 215123, China
12 b Department of Environmental Science, University of Liverpool, Brownlow Hill, Liverpool L69
13 7ZX, United Kingdom
14 c Department of Biological Sciences, Xi’an Jiaotong-Liverpool University, Suzhou 215123,
15 China
16 * Corresponding author. Department of Environmental Science, Xi’an Jiaotong-Liverpool
17 University, Suzhou, Jiangsu 215123, China.
18 E-mail address: [email protected] or [email protected] (Z. Chen)
19
© 2018 by the author(s). Distributed under a Creative Commons CC BY license.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 8 February 2018 doi:10.20944/preprints201802.0067.v1
Peer-reviewed version available at Journal of Soils and Sediments 2018; doi:10.1007/s11368-018-2024-1
Page 2 of 26

20 Abstract: Soil microbial fuel cells (sMFC) are a novel technique that use organic matters in soils
21 as an alternative energy source. External resistance (ER) is a key factor influencing sMFC
22 performance and, furthermore, alters the soil’s biological and chemical reactions. However, little
23 information is available on how the microbial community and soil component changes in sMFC
24 with different ER. Therefore, the effects of anodes of sMFC at different ER (2000 Ω, 1000 Ω, 200
25 Ω, 80 Ω and 50 Ω) were examined by measuring organic matter (OM) removal efficiency, trace
26 elements in porewater and bacterial community structure in contaminated paddy soil. The results
27 indicated that ER has significant effects on sMFC power production, OM removal efficiency and
28 bacterial beta diversity. Moreover ER influences iron, arsenic and nickel concentration as well in
29 soil porewater. In particular, greater current densities were observed at lower ER (2.4mA, 50Ω)
30 compared to a higher ER (0.3mA, 2000Ω). The removal efficiency of OM increased with
31 decreasing ER whereas it decreased with soil distance away from the anode. Furthermore, principal
32 coordinate analysis (PCoA) revealed that ER may shape the bacterial communities that develop in
33 the anode vicinity but have minimal effect on that of the bulk soil. The current study illustrates
34 that lower ER can be used to selectively enhance the relative abundance of electrogenic bacteria
35 and lead to high OM removal.
36 Keywords: External resistances; Soil microbial fuel cells; Paddy soil; Geobacter; Arsenic; Iron;
37 Organic matter
38
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 8 February 2018 doi:10.20944/preprints201802.0067.v1
Peer-reviewed version available at Journal of Soils and Sediments 2018; doi:10.1007/s11368-018-2024-1
Page 3 of 26

39 1. Introduction
40 Organic carbon in sediments or soils can be used as a new source of energy when employed
41 in microbial fuel cells (MFC). Soil microbial fuel cells (sMFC) are a special type of MFC that
42 produce electricity by using organic chemicals present in soils as energy source [1-4]. In sMFC
43 organic matter (OM) is oxidized at the anode by the soil microorganism, resulting in the production
44 of electrons. The electrons produced during OM oxidation migrate towards the cathode, through
45 an external circuit, where they combine with oxygen and protons to form water. sMFC that employ
46 paddy soil has gained considerable attention recently because of the unique biological and
47 chemical characteristics of paddy soil make them ideal for power production [5]. In addition to
48 power generation, sMFC have also found use in stimulating soil bioremediation. For instance, the
49 anode of the sMFC can serve as an electron sink for anode respiring bacterial (ARB) and accelerate
50 the bioremediation of both polycyclic aromatic hydrocarbons and redox active heavy metals [6,7].
51 However, sMFC deployment in paddy fields has been shown to alter soil bacterial
52 community structure and abiotic components [8,9]. When an anode of sMFC is embedded into
53 anaerobic soils, it functions as a continuous sink for electrons and this alters soil chemistry [10,11].
54 This in turn increases soil Eh, decreases soil pH and changes the bioavailability of soil nutrients
55 such as phosphate and nitrogen [1,12]. Therefore these changes in soil chemistry may influence
56 the soil bacterial community composition, since soil Eh, pH and nutrient bioavailability are the
57 major drivers of soil bacterial community structure [13-15]. Furthermore, the sMFC have also been
58 used to prevent the deterioration of overlying water quality and for the removal or stabilization of
59 toxic heavy metals [7,16,17]. Previous studies have shown that sMFC can be used as a mitigation
60 technology in soils contaminated with chromium, uranium, cadmium, lead or copper [7,17,18].
61 Anode bacterial community determines the capability of the sMFC to function effectively
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 8 February 2018 doi:10.20944/preprints201802.0067.v1
Peer-reviewed version available at Journal of Soils and Sediments 2018; doi:10.1007/s11368-018-2024-1
Page 4 of 26

62 as a remediation technology. For example, when non-anode respiring bacteria are abundant in the
63 anode chamber, these microbes may compete with ARB for resources, thereby limiting the
64 performance of the sMFC. However, regulating the anode potential has been shown to effectively
65 tailor the desired ARB consortium and suppress the growth of undesirable microbes, such as
66 methanogens [9,19]. This occurs because the anode potentials determine the accessibility of
67 electron ARB received for growth and development [20]. Thus at certain anode potential, ARB
68 are able to gain higher energy for growth from respiring the anode and out compete methanogens
69 for OM [19].
70 Moreover, adjusting applied external resistance (ER) in sMFC has also been shown to
71 affect the metabolic rate of ARB on the anode [21]. Previous studies have demonstrated that
72 decreasing ER increases the oxidative degradation of OM and the current output from the sMFC
73 [3]. Furthermore, the current produced by the sMFC can also stimulate microbiological oxidation
74 of soil OM by fermentative bacteria [6,22]. Even though, several studies have examined the
75 influence of the anode potentials on the biofilm community composition in double chamber MFC,
76 to date there is no universal consensus on the effect of the anode potential on the anode community
77 structure. Although the application of sMFC is increasing, studies examining the effect of anode
78 potential on sMFC community have received limited to no attention. Torres, et al. [20], observed
79 selective enhancement of Geobacter sulfurreducens at lower anode potentials (−0.15V, −0.09V,
80 and 0.02V vs SHE) and an increase in bacterial diversity at higher anode potential (0.37V). On the
81 contrary, Zhu, et al. [23] argues that anode community is unaffected by anode potential and that
82 the electrogenic communities acclimatize to different anode potentials. Nonetheless, to date, few
83 studies have been done to examine the influence of the anode at different ER on soil biotic and
84 abiotic constituents collectively.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 8 February 2018 doi:10.20944/preprints201802.0067.v1
Peer-reviewed version available at Journal of Soils and Sediments 2018; doi:10.1007/s11368-018-2024-1
Page 5 of 26

85 Thus, the objectives of this study are, therefore, to examine the effects of different ER on
86 soil and anode bacterial community structure, OM removal efficiency and selective soil metal
87 behavior at various resistances away from the anode. The results from this study suggest that lower
88 ER can be used to selectively enhance ARB abundance and increase OM removal.
89 2. Materials and Methods
90 2.1. Paddy soils sample
91 The paddy soil (0–15 cm below the soil–water interface) samples were collected from a
92 rice paddy in Qiyang, Hunan (GPS N26.760 E111.86) and transported directly to the laboratory.
93 These samples were then air-dried and sieved to 2 mm to remove all coarse debris. The soil
94 properties, including texture, pH, OM content, arsenic (As) and Iron (Fe) were measured to be clay,
95 6, 23.2g/kg, 73.7mg/kg and 53.69 g/kg, respectively.
96 2.2. Soil Microbial Fuel Cell assembly
97 Eighteen sMFC were constructed from the paddy soil and operated at different ER 2000 Ω,
98 1000 Ω, 200 Ω, 80 Ω and 50 Ω in triplicates for 90 days. Of the eighteen sMFC three were controls,
99 the anodes and cathodes were not connected (open circuit). All sMFC were constructed according
100 to a previously reported method by Wang, et al. [24] with slight modifications. Briefly, a columnar
101 polyethylene terephthalate container (10 cm diameter × 15 cm depth) with two valve ports (~2 cm
102 above the anode and adjacent to the anode) was used to construct each sMFC. Circular carbon felts
103 with geometric surface area of 50.2 cm2 were be used as anodes and cathodes. A data logger (USB-
104 7660B, ZTIC, China) was used to record the voltage between the anode and cathode.
105 Using 700 g (dry weight) of soil sample, ~1 cm depth of soil was placed at the bottom of
106 the sMFC container. Then, the anode was placed on the surface of the soil layer and then buried
107 with additional soil to simulate anaerobic conditions. The cathode was placed above the soil in
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 8 February 2018 doi:10.20944/preprints201802.0067.v1
Peer-reviewed version available at Journal of Soils and Sediments 2018; doi:10.1007/s11368-018-2024-1
Page 6 of 26

108 aerobic conditions (half submerged in water and the other half in the open air). Deionized water
109 (1L) was added to flood the paddy soil.
110 2.3. Chemical Analysis
111 Each constructed sMFC was equipped with a self-made soil pore water sampler. The
112 sampler was made from a 0.45 µm hollow fiber membrane, which was inserted into the soil
113 through the valve port adjacent to the anode and 2 cm above the anode. The soil porewater was
114 sampled and analyzed for As and Fe by inductively coupled plasma emission spectrometry
115 (ICPMS) (NexlonTM 350x, Pekin Elmer, USA) and atomic absorption spectrometry (AAS)
116 (PinAAcle™ 900, PerkinElmer, USA), respectively. Dissolved organic carbon (DOC)
117 concentration was determined with a TOC analyzer (Shimadzu TOC-VCPH, Japan). The loss on
118 ignition (LOI) carbon content was determined gravimetrically. LOI was calculated according to
119 the following equation 1:
LOI (%) = ((D105 –D550)/D105) *100 (1)
120
121 where D105 is the dry mass of the sample heated 105 ºC for 12 hours and D550 is the dry mass of
122 the sample after 4 hours of incubation at 550 ºC. Samples were allowed to cool to room temperature
123 in a desiccator before determining D550.
124 2.4. Bacterial Community Analysis
125 Total Genomic DNA was extracted from the anode vicinity and bulk soil using the
126 PowerSoil DNA isolation kit (Mo Bio Laboratories, Inc., Carlsbad, CA) from all of the treatments
127 including the controls according to the manufacturer’s protocol. The quantity of DNA in the
128 extracts was measured at 260 nm using a Qubit 2.0 Fluorometer (Invitrogen, Carlsbad, CA). The
129 DNA quality was verified by agarose gel electrophoresis. The DNA samples were stored at -80 °C
130 until sequencing.
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 8 February 2018 doi:10.20944/preprints201802.0067.v1
Peer-reviewed version available at Journal of Soils and Sediments 2018; doi:10.1007/s11368-018-2024-1
Page 7 of 26

131 The V3-V5 hypervariable regions of the 16S ribosomal RNA (rRNA) gene in bacteria were
132 amplified using primer pair 515f/907R [25,26]. The library preparations and Illumina MiSeq
133 sequencing of the 16S rRNA gene amplicon were outsourced to GENEWIZ, Inc. (Suzhou, China).
134 FLASH was used to merge paired-end reads and QIIME (Quantitative Insights Into Microbial
135 Ecology) (V1.7.0) was employed to optimize the data by removing low-quality sequence and for
136 all taxonomy analysis [27]. The low-quality reads were discarded. Subsequently, the sequences
137 were compared with the reference database (RDP Gold database) using UCHIME algorithm to
138 detect chimeric sequence, and then the chimeric sequences were removed [28,29]. The clean
139 sequences were clustered into operational taxonomic units (OTUs) (97% similarity) by UPARSE.
140 The phylogenetic taxonomy were assigned according to the Ribosomal Database Program (RDP)
141 classifier (Version 2.2) and the Green Genes Database at confidence threshold of 80% [28,30].
142 2.5. Statistical Analysis
143 One-way analysis of variance (ANOVA) was performed to test significant differences
144 between treatments using SPSS software (IBM SPSS Statistics 23.0) and OriginPro (Inc.,
145 OringinLab, USA). Tukey’s honest significant difference (HSD) studentized range tests were
146 applied to differentiate between various treatments and a p < 0.05 was considered significant.
147 Alpha diversity indexes were calculated using Faith’s phylogenetic diversity, Shannon diversity
148 index and the Chao1 richness estimator using QIIME. Beta diversity was assessed using principal
149 coordinate analysis (PCoA) and non-metric multidimensional scaling (NMDS) analysis.
150 2.6 Nucleotide sequence accession number
151 Nucleotide sequences were all deposited in the GeneBank database with accession numbers
152 MG814044 - MG815131.
153
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 8 February 2018 doi:10.20944/preprints201802.0067.v1
Peer-reviewed version available at Journal of Soils and Sediments 2018; doi:10.1007/s11368-018-2024-1
Page 8 of 26

154 3. Results
155 3.1. Electricity generation from soil MFCs with different external resistors
156 In this study each sMFC was loaded with a fixed resistor for the duration of the experiment.
157 Five different ER were tested 50 Ω, 80 Ω, 200 Ω, 1000 Ω and 2000 Ω and the current production
158 from the sMFC for 90 days of operation are illustrated in Figure 1. The current of the all sMFC
159 sharply increased during the initial stage regardless of ER and reached a maximum current at
160 around 10 days. The highest current produced was observed at an ER of 50 Ω (2.4mA), followed
161 by those of 80 Ω (2.1 mA), 200 Ω (1.6 mA) and 1000 Ω (0.8 mA). An ER of 2000 Ω produced the
162 lowest current (0.3mA). However after ca. 30 days all the current outputs irrespective of ER were
163 approximately the same (0.15-0.2 mA). The insert in Figure 1 shows change in the controls voltage
164 with time.
165
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 8 February 2018 doi:10.20944/preprints201802.0067.v1
Peer-reviewed version available at Journal of Soils and Sediments 2018; doi:10.1007/s11368-018-2024-1
Page 9 of 26

166 Figure 1. Current production from sMFC fixed with different external resistances during a 90-
167 day operation. The insert represent the change in the control’s voltage with time.
168 3.2. Polarization properties of the sMFC
169 The power current curve is an important attribute used to determine the performance of
170 MFC. The power current curves in this study were obtained by varying ER from 10KΩ to 10Ω
171 and are compared in Figure 2. The power densities of the sMFC set at 1000 Ω and 2000 Ω ER
172 were similar and those set at lower ER (50 Ω, 80 Ω and 200 Ω) mirrored each other. As shown in
173 Figure 2, the maximum power density observed in the sMFC at different resistance were 22.2,
174 22.2, 24.1, 36.5 and 40.0 mW/m2 for 50 Ω, 80 Ω, 200 Ω, 1000 Ω and 2000 Ω respectively. Similarly,
175 the polarization slope method was used to determine the ohmic resistance for each sMFC. The
176 results followed a similar trend to maximum power except for the case of 200 Ω. The 50 Ω ER
177 had the lowest ohmic resistance and 2000 Ω had the highest ohmic resistance. The ohmic resistance
178 increased in the following order 50 (389 Ω) < 80 (390 Ω) < 1000 (476 Ω) < 200 (556 Ω) < 2000
179 (572 Ω).
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 8 February 2018 doi:10.20944/preprints201802.0067.v1
Peer-reviewed version available at Journal of Soils and Sediments 2018; doi:10.1007/s11368-018-2024-1
Page 10 of 26

180
181 Figure 2. Power density (filled symbols) and polarization curves (open symbols) of sMFCs. The
182 error bars represent standard error of measured concentrations of triplicate samples.
183 The anode and cathode potentials (versus Ag/AgCl reference electrode) for each sMFC at
184 different ER are shown in Figure S1. In general, the anode potential became more positive with
185 decreasing ER, however there was negligible difference between the potential of the anode at 50
186 Ω, 80 Ω and 200 Ω. The working potential of the anode for the sMFC with an ER of 2000 Ω was
187 much lower than those with other ER, approximately 345 mv lower than those with ER of 50 Ω,
188 80 Ω and 200 Ω. Moreover, the highest cathodic potential was observed in the sMFC with 200 Ω
189 ER followed by that of 80 Ω and 1000 Ω. The cathodic potential of sMFC with ER of 50 Ω and
190 2000 Ω were approximately the same. These findings indicate that the cathode was responsible for
191 the difference in sMFC performance with different ER, especially at ER of 200 Ω, 80 Ω and 50 Ω